Safe and Scalable Web Agent Learning via Recreated Websites
This addresses the challenge of unsafe and limited training environments for web agents, offering a scalable solution for autonomous web interaction.
The authors tackled the problem of training autonomous web agents by proposing VeriEnv, a framework that clones real-world websites into synthetic environments, enabling agents to self-generate tasks with verifiable rewards, which resulted in agents generalizing to unseen websites and achieving site-specific mastery through scalable training.
Training autonomous web agents is fundamentally limited by the environments they learn from: real-world websites are unsafe to explore, hard to reset, and rarely provide verifiable feedback. We propose VeriEnv, a framework that treats language models as environment creators, automatically cloning real-world websites into fully executable, verifiable synthetic environments. By exposing controlled internal access via a Python SDK, VeriEnv enables agents to self-generate tasks with deterministic, programmatically verifiable rewards, eliminating reliance on heuristic or LLM-based judges. This design decouples agent learning from unsafe real-world interaction while enabling scalable self-evolution through environment expansion. Through experiments on web agent benchmarks, we show that agents trained with VeriEnv generalize to unseen websites, achieve site-specific mastery through self-evolving training, and benefit from scaling the number of training environments. Code and resources will be released at https://github.com/kyle8581/VeriEnv upon acceptance.